Fibrosing interstitial lung disease (fILD) refers to a group of lung diseases that result in significant morbidity and mortality due to progressive scarring of the alveolar interstitium. There are no effective treatments to reliably relieve symptoms or prolong life. FILD is a complex disease likely caused by the interplay among multiple genetic and environmental factors, most of which are unknown. This project is motivated by two large genome-wide studies of fILD that we are completing: one familial linkage study and one case-control association study. We will sequence cases from the familial study in addition to other cases and controls to identify the risk loci within the high priority regions fro these studies. A priority for the sequencing study will be rare (<1%) and uncommon (<5%) variants. Several tests of association for rare and uncommon variants have been developed in the case-control setting, where the variants are aggregated to improve statistical power. However, none of these statistical tests use family-specific linkage information, information that is highly informative for genetic heterogeneity. Hence, there are no existing methods for integration of family-specific linkage information into aggregate tests of association. The goal of this project is to fill this gap in statistical methodology for directly combining family-specific linkage information with other case-control sequence data in order to identify genetic risk variants for fILD. The central hypothesis of this project is that the power of genetic association tests for rare variants can be improved by using family-specific allele sharing information to give more weight to familial cases from families with evidence for linkage at the locus of interest. We propose extensions of rare/uncommon variant tests of association that a) require only one case per family, b) use external measures of allele sharing from linkage studies, and c) allow joint analysis of familial cases with linkage information and other cases. The first aim is to evaluate a new framework for weighted tests of association between a group of rare/uncommon genetic variants and a dichotomous trait. The contribution of each familial case is weighted according to the case's family-specific evidence for allele sharing at the locus of interest. The second aim is to apply the new methods to fILD to identify novel variants that increase the risk of fILD and compare the results to those obtained using existing methods. The third aim is to develop and disseminate a freely-available implementation of the tests developed in aim 1, which will facilitate the application of these methods to other disorders. While two studies of fILD motivate this work, there are hundreds of similar studies with available linkage data which would benefit from the increased power to detect risk variants with these tests. The successful completion of this science will result in tests of association and software applicable to resequencing studies that improve power by leveraging the information contained in, and the resources that have been devoted to, familial linkage cohorts. The methods are naturally extendable to quantitative traits and other complex testing strategies to increase power to detect functional variants.